Slightly off topic but hopefully useful...

Just a little bit of trivia with the Intel Compilers on Mac OS X. One of the nuances about the Intel compilers, unlike GCC, are that they are two different compilers, a 64-bit compiler and a 32-bit compiler. The 64-bit compilers are installed under the CCE/FCE directory and the 32-bit compilers are CC/FC. Prior to v10.1 the only way to switch between the 32-bit and 64-bit compilers was to run the supplied shell script to set the environment variables. However with v10.1, one can now execute the -m32 or -m64 flag for the Intel compilers as well, this makes it much easier to work with 32-bit or 64-bit builds!



Warner Yuen

Scientific Computing 

Consulting Engineer

Apple, Inc.

email: wyuen@apple.com


On Apr 25, 2008, at 9:00 AM, users-request@open-mpi.org wrote:

Message: 1
Date: Fri, 25 Apr 2008 07:08:37 -0400
From: Jeff Squyres <jsquyres@cisco.com>
Subject: Re: [OMPI users] install intel mac with Laopard
To: Open MPI Users <users@open-mpi.org>
Message-ID: <4559DCD6-4C21-44C2-BAE8-A711404EAFE9@cisco.com>
Content-Type: text/plain; charset=US-ASCII; format=flowed

Sorry, I should have been more specific: how about this?

**********************************************************************
It appears that your Fortran 77 compiler is unable to link against
object files created by your C compiler.  This typically indicates
one of a few possibilities:

  - A conflict between CFLAGS and FFLAGS
  - A problem with your compiler installation(s)
  - Different default build options between compilers (e.g., C
    building for 32 bit and Fortran building for 64 bit)
  - Incompatible compilers

Such problems can usually be solved by picking compatible compilers
and/or CFLAGS and FFLAGS.  More information (including exactly what
command was given to the compilers and what error resulted when the
commands were executed) is available in the config.log file in this
directory.
**********************************************************************

On Apr 25, 2008, at 7:00 AM, Jeff Squyres wrote:

How about a compromise -- I'll extend the message to also include the
possibility of architecture mismatches.


-- 
Jeff Squyres
Cisco Systems