This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: RFC POSIX Fortran Interface
- From: Walter Spector <w6ws at earthlink dot net>
- To: Fortran List <fortran at gcc dot gnu dot org>
- Date: Sun, 14 Oct 2007 14:40:17 -0700
- Subject: Re: RFC POSIX Fortran Interface
- Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=dk20050327; d=earthlink.net; b=EvRRuwgVrfbi5Lrl3qgtuAFXReS0rOBduUhECzBb3Vnlj9ruGgFzHUBhc0RCdL4h; h=Received:Message-ID:Date:From:Organization:X-Mailer:X-Accept-Language:MIME-Version:To:Subject:References:Content-Type:Content-Transfer-Encoding:X-ELNK-Trace:X-Originating-IP;
- Organization: Not much.
- References: <471267C1.2000907@verizon.net> <20071014201119.GA64664@troutmask.apl.washington.edu> <47127F05.4060000@verizon.net>
Jerry DeLisle wrote:
> ...
> Ideally, to me anyway, a user ought to be able to just have "use pxf_interface
> in their own code and have the compiler just do it without need for a compiler
> option (an intrinsic module)...
I would argue against this for two reasons:
1.) POSIX-1003.9 was written against Fortran-77, where modules did not exist.
Therefore we should not require that a module be USEd in order to use the routines.
2.) POSIX-1003.9 failed to provide a suggested name for a Fortran-90 module.
So while a module is extremely useful to help the programmer get the calls right
(and perhaps use keyword=value argument style), there is no consensus on what
the correct name for the module should be. For example, in the Intel compiler,
Intel supplies a module called IFPOSIX. On IRIX, Cray, and the Open64 compilers,
the module is called PXF_DEFINITIONS... (I should note that in the latter case,
the library was written by a member of the 1003.9 committee.)
Walter