From owner-sc22wg5@open-std.org Tue Dec 9 21:47:18 2008 Return-Path: X-Original-To: sc22wg5-dom7 Delivered-To: sc22wg5-dom7@www2.open-std.org Received: by www2.open-std.org (Postfix, from userid 521) id 6C103CA343D; Tue, 9 Dec 2008 21:47:18 +0100 (CET) X-Original-To: sc22wg5@open-std.org Delivered-To: sc22wg5@open-std.org Received: from nspiron-2.llnl.gov (nspiron-2.llnl.gov [128.115.41.82]) by www2.open-std.org (Postfix) with ESMTP id 7762BCA3439 for ; Tue, 9 Dec 2008 21:47:16 +0100 (CET) X-Attachments: None Received: from cyrus2.llnl.gov ([128.15.97.105]) by nspiron-2.llnl.gov with ESMTP; 09 Dec 2008 12:47:15 -0800 From: Aleksandar Donev Organization: LLNL To: WG5 Subject: Re: (j3.2006) (SC22WG5.3802) [ukfortran] N1761, TYPE(*), BIND(C) and arrays Date: Tue, 9 Dec 2008 12:47:15 -0800 User-Agent: KMail/1.9.4 References: <20081127193527.EF00DC178D9@www2.open-std.org> <20081209185450.B4B95CA3439@www2.open-std.org> <20081209203233.1E8AFCA3439@www2.open-std.org> In-Reply-To: <20081209203233.1E8AFCA3439@www2.open-std.org> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Message-Id: <200812091247.15467.donev1@llnl.gov> Sender: owner-sc22wg5@open-std.org Precedence: bulk On Tuesday 09 December 2008 12:32, N.M. Maclaren wrote: > Bill keeps talking C calling Fortran. =A0Fine. =A0In THAT case, I agree > that all of the arguments are dummy arguments. =A0That isn't the issue. > =A0What about the case of Fortran calling C? =A0PLEASE take a look at > that program and consider the issues. What issues Nick? You sent an example program, which seemed perfectly=20 harmless to me, and I said that. Sorry I cannot read minds. There is no difference between Fortran calling a Fortran BIND(C)=20 procedure and Fortran calling a C procedure. The compiler does exactly=20 the same thing in both cases, and it need not do any type of analysis=20 to figure out what the other procedure is written in. Exactly the same=20 restrictions on actual arguments apply (e.g., non-aliasing rules,=20 pointer association preservation rules, etc.). This is exactly how it=20 is now in C Interop. In all of your examples, just write the procedure Slough in Fortran: =A0 =A0 =A0 =A0 SUBROUTINE Slough (arg) BIND(C) =A0 =A0 =A0 =A0 =A0 =A0 TYPE(*), DIMENSION(..) :: arg WRITE(*,*) "RANK=3D", RANK(arg) =A0 =A0 =A0 =A0 END SUBROUTINE Slough Is there a problem then? What is this problem, please explain. The=20 interoperable C interface is: void slough(void * f_desc); > you and Bill seem to do is to deny that problems exist, even if they > are shown to exist (and be serious). No, Reinhold, for example, brought some excellent points and I agree=20 with most of them, and have modified my own ballot (yet to be finished)=20 accordingly. You also made some good points (e.g., flexible array=20 members) in your comments, which I agree with and am ammending. What I=20 disagree with is your outrageous comment "this makes no sense, your=20 design is flawed from the basics, etc. etc." Sorry, it made sense to at=20 least 4 of us involved in writing it. Best, Aleks =2D-=20 Aleksandar Donev, Ph.D. Lawrence Postdoctoral Fellow @ Lawrence Livermore National Laboratory High Performance Computational Materials Science and Chemistry E-mail: donev1@llnl.gov Phone: (925) 424-6816 Fax: (925) 423-0785 Address: P.O.Box 808, L-367, Livermore, CA 94551-9900 Web: http://cherrypit.princeton.edu/donev