[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

20031212: netCDF beta with ifc and icc

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: 20031212: netCDF beta with ifc and icc
  • Date: Fri, 12 Dec 2003 13:19:27 -0700
Timothy,

> To: address@hidden
> From: Timothy J Campbell <address@hidden>
> Subject: netCDF beta with ifc and icc
> Organization: Mississippi State/Stennis Space Center
> Keywords: netCDF Linux Intel compilers

The above message contained the following:

> I've searched through the support archives and haven't found anything 
> that directly addresses what I am experiencing.
> 
> I am trying to install netCDF 3.5.1-beta13 on Intel Linux using the 
> Intel Fortran and C version 7.1 compilers.  I am able to configure and 
> compile without errors (although there are a few strange warnings). 
> However, the "make test" has 36 failures that are all related to getting 
> and putting floats.  Attached are log files from the configure, make, 
> and "make test" steps.
> 
> Here is some system and version info:
> 
> %> uname -srvmpio
> Linux 2.4.20-20.8smp #1 SMP Mon Aug 18 14:39:22 EDT 2003 i686 i686 i386 
> GNU/Linux
> 
> %> ifc -V
> Intel(R) Fortran Compiler for 32-bit applications, Version 7.1   Build 
> 20030307Z
> 
> %> icc -V
> Intel(R) C++ Compiler for 32-bit applications, Version 7.1   Build 20030307Z
> 
> I found that the cfortran.h was identical to that of version 3.5.0 -- so 
> I applied the intel-f95-patch from Carlie Coats.  Here is the 
> environment I set before configuring:
> 
> export CC=icc
> export CXX=icc
> export FC=ifc
> export F90=ifc
> export CPPFLAGS="-DNDEBUG -DINTEL_COMPILER"
> export FFLAGS="O -mp -cm -w95"
> export F90FLAGS="O -mp -cm -w95"

I hope the above "O"s are typos (they should be "-O").

> export CFLAGS=-O
> export FLIBS=-lPEPCF90
> 
> Here are the configure and make steps I followed.
> 
> %> cd /home/tjcamp/netcdf
> %> ./configure --prefix=/home/tjcamp/netcdf >& configure.log
> %> make >& make.log
> %> make test >& test.log
> 
> If there is any other information I should provide -- let me know. 
> Thank-you in advance for any help you may provide.

We don't have your system here, so we can't duplicate your problem.
According to others with your type of system, however, you have the
following options:

    
http://my.unidata.ucar.edu/content/software/netcdf/other-builds.html#intel-ifc

    
http://my.unidata.ucar.edu/content/software/netcdf/other-builds.html#intel-ifc 
3.5.1-beta13

I suggest that you experiment with adding either "-Vaxlib" or
"-dynamic-libcxa" to the FLIBS environment variable before executing the
configure script.  Don't forget to do a "make distclean" before
executing the configure script.

Please let me know what you discover.

> Tim
> -- 
> Timothy J. Campbell, Ph.D.
> Mississippi State University   | Ph: 228-688-3518
> 1103 Balch Blvd, Rm 248        | Fx: 228-688-2764
> Stennis Space Center, MS 39529 | address@hidden

Regards,
Steve Emmerson