[4035] in RedHat Linux List
Re: Fortran 77 and 90
daemon@ATHENA.MIT.EDU (Joe Louderback)
Tue Nov 12 09:46:59 1996
From: "Joe Louderback" <jlouder@wfu.edu>
To: redhat-list@redhat.com
Date: Tue, 12 Nov 1996 09:45:13 +0000
Reply-To: jlouder@wfu.edu
Resent-From: redhat-list@redhat.com
Michael K Johnson wrote
> Consider that most of the respected commercial Fortran compilers
> convert the fortran code to C, and the use a C compiler to compile
> the C, and then f2c doesn't seem like such a bad idea. I am told
> (I am, thankfully, not a fortran user myself) that it works rather
> well.
Respectfully sir, I beg to differ on your first point. Some
compilers do first translate to C -- I believe NAG F90 did so.
However, many do not. They may translate to the same intermediate
language that the company's C compiler does. This allows the same
back end to be used for both products.
A reason for not using C as an intermediary is efficiency. Fortran
compilers can generally perform more aggressive optimizations , and
this is lessened if C is used (g77 apparently produces faster code
than f2c +gcc). There is a price for flexibility ;-)
> g77 has certainly matured lately, but I seem to recall a need for
> Fortran 90 in one of the recent messages. *Every* Fortran 90
> compiler for Unix that I'm aware of compiles to C and then
> compiles the C into object code.
DEC F90 does not. NAG now has a native F90 compiler, I believe.
F90 to C translators were the rule for a while, but native compilers
are now available.
Joe Louderback, who spends far too much time hanging out in
comp.lang.fortran
--
PLEASE read the Red Hat FAQ, Tips, Errata and the MAILING LIST ARCHIVES!
________________________________________________________________________
http://www.redhat.com/RedHat-FAQ http://www.redhat.com/RedHat-Errata
http://www.redhat.com/RedHat-Tips http://www.redhat.com/mailing-lists
------------------------------------------------------------------------
To unsubscribe: mail -s unsubscribe redhat-list-request@redhat.com < /dev/null
