Date: Wed, 10 Jul 2013 10:07:11 -0700 From: Steve Kargl <sgk@troutmask.apl.washington.edu> To: mexas@bristol.ac.uk Cc: freebsd-fortran@freebsd.org Subject: Re: why fortran mailing list? Message-ID: <20130710170711.GA59476@troutmask.apl.washington.edu> In-Reply-To: <201307100759.r6A7xqsK098965@mech-cluster241.men.bris.ac.uk> References: <20130709231808.GA92444@troutmask.apl.washington.edu> <201307100759.r6A7xqsK098965@mech-cluster241.men.bris.ac.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Jul 10, 2013 at 08:59:52AM +0100, Anton Shterenlikht wrote: > From sgk@troutmask.apl.washington.edu Wed Jul 10 00:33:34 2013 > >> OpenMPI is >> the only Fortran code that I routinely build outside of the >> ports. > > Please elaborate. > I don't know much about the differences between OpenMPI > and mpich. The latter seems to work: > > # pkg info -xr mpi > mpich2-1.3.2.p1_1,5: > scotch-5.1.12.b.e > astk-serveur-1.11.0 > blocksolve95-3.0_10 > ParMetis-4.0_1 I've used both mpich2 and OpenMPI on i386 and amd64 systems. In the testing I've done with my applications, OpenMPI simply out performed mpich2. This was a while ago, so things may have changed. OpenMPI just works. The one issue I had with OpenMPI was resolved within a day. Note, I do not use ports/net/openmpi. I download the latest version from the OpenMPI, unpack it, run configure and install. -- Steve
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20130710170711.GA59476>