GFortran 44 optimizations and Intel SSE4

[Jacob Anderson]

For our MVAPICH2 (1.7) build, I am using this command line on the E5620 and X5670 machines:

./configure --prefix=/usr/mpi/gcc/mvapich2-1.7 
  --enable-fast=O3 
  MPICH2LIB_CFLAGS='-march=native -msse4.2 -maes' 
  MPICH2LIB_FFLAGS='-march=native -msse4.2 -maes'
  MPICH2LIB_CXXFLAGS='-march=native -msse4.2 -maes' 
  MPICH2LIB_FCFLAGS='-march=native -msse4.2 -maes' 
  --with-file-system=nfs 
  GCC=/usr/bin/gcc44 
  CC=/usr/bin/gcc44 
  CXX=/usr/bin/g++44 
  F77=/usr/bin/gfortran44 
  FC=/usr/bin/gfortran44

This appeared to build just fine without any warnings that I saw.



> -----Original Message-----
> From: fortran-owner@gcc.gnu.org [mailto:fortran-owner@gcc.gnu.org] On
> Behalf Of Tim Prince
> Sent: Tuesday, November 01, 2011 10:19 AM
> To: fortran@gcc.gnu.org
> Subject: Re: GFortran 44 optimizations and Intel SSE4
> 
> On 11/1/2011 12:42 PM, Jacob Anderson wrote:
> >
> > Thanks for the responses! Yes, I did mean the "-Ox" generic options.
> >
> > We are compiling against Intel X5670 and X5620, so no desktop
> architectures.  The "SSE4.2" extensions are of interest for their vector
> capabilities. Is there any online documentation on the optimization
> algorithms used when the SSE4 optimizations are enabled?
> >
> > We are using gfortran44 version 4.4.4 20100726:
> >
> 
> X86_64 generic options include auto-vectorization, which might
> occasionally be improved by adding -msse4.  i386 generic options do not;
> you would require an option such as -march=core2.
> 
> 
> --
> Tim Prince