Date: Wed, 10 Jul 2013 10:16:36 -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: <20130710171636.GB59476@troutmask.apl.washington.edu> In-Reply-To: <201307100820.r6A8KkNm099328@mech-cluster241.men.bris.ac.uk> References: <20130709233149.GA93320@troutmask.apl.washington.edu> <201307100820.r6A8KkNm099328@mech-cluster241.men.bris.ac.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Jul 10, 2013 at 09:20:46AM +0100, Anton Shterenlikht wrote: > From sgk@troutmask.apl.washington.edu Wed Jul 10 00:33:37 2013 > > On Tue, Jul 09, 2013 at 07:24:12PM -0400, Christopher J. Umina wrote: > > Fortran is pretty much a standard in the scientific world and I for > > one have been quite upset about the fact that I'm constantly forced > > away from FreeBSD for such things. Needless to say I was quite > > surprised that a list was created specifically for Fortran discussion, > > but I'm also very excited to see the level of interest in the > > community. > > > > Please, don't top-post, it looses context. In fact, I've > deleted all of the previous context. :-) > > Not sure why you're forced away from FreeBSD because of > Fortran. gfortran has been a part of GCC since 2005. > Although the first year or two of gfortran's life saw a > large number of bugsi and patches, since gcc 4.4,x ori > 4.5.x, she's been a very good compiler. > > Right now I need coarrays, so I have to use other compilers > on linux. gfortran has single-image co-array support. As you know, Tobias has slowly been working on co-arrays, and IIRC the only missing functionality is runtime support. He could use some help writing that support, so now is your chance to contribute. > But a compiler alone is not enough. > I need a complete tightly integrated fortran > environment, with libraries and applications. Then, you're on your own. The FreeBSD ports collection cannot provide the tight integration that you desire. > Many libs/apps available on linux are not > in ports yet. One example is Trilinos, > http://trilinos.sandia.gov/packages/fortrilinos/. > I think bf@ has been working on a port. > But there are more, I just can't think of > any at the moment. Why wait for a port to appear in the Port Collection. Grab the code you need and build it. > Then there are things like legacy code. > For example, I myself maintain math/slatec, > which is 1980-1990 fortran 77 library > with over 1400 subroutines. > I found this library useful at some point. > But the code is old and completely unmaintained, > perhaps there are better newer alternatives. > Should it stay in ports for now? Of course. Why remove a functioning port? The beauty of old (standard conforming) Fortran coder is that it will compile with modern Fortran compilers. -- Steve
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20130710171636.GB59476>