From owner-sc22wg5@open-std.org Mon Dec 8 16:42:07 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 7C3E0C178E5; Mon, 8 Dec 2008 16:42:07 +0100 (CET) X-Original-To: sc22wg5@open-std.org Delivered-To: sc22wg5@open-std.org X-Greylist: delayed 949 seconds by postgrey-1.18 at www2.open-std.org; Mon, 08 Dec 2008 16:42:06 CET Received: from proofpoint3.lanl.gov (proofpoint3.lanl.gov [204.121.3.28]) by www2.open-std.org (Postfix) with ESMTP id 96DAFC178E0 for ; Mon, 8 Dec 2008 16:42:05 +0100 (CET) Received: from mailrelay2.lanl.gov (mailrelay2.lanl.gov [128.165.4.103]) by proofpoint3.lanl.gov (8.13.8/8.13.8) with ESMTP id mB8FQDtr028195 for ; Mon, 8 Dec 2008 08:26:13 -0700 Received: from localhost (localhost.localdomain [127.0.0.1]) by mailrelay2.lanl.gov (Postfix) with ESMTP id 93F5615D9807 for ; Mon, 8 Dec 2008 08:26:13 -0700 (MST) X-NIE-2-Virus-Scanner: amavisd-new at mailrelay2.lanl.gov Received: from cic-mail.lanl.gov (cic-mail.lanl.gov [128.165.4.115]) by mailrelay2.lanl.gov (Postfix) with ESMTP id 6706315D97FC for ; Mon, 8 Dec 2008 08:26:13 -0700 (MST) Received: from soren.lanl.gov (soren.lanl.gov [128.165.148.20]) by cic-mail.lanl.gov (Postfix) with ESMTP id D5419208007 for ; Mon, 8 Dec 2008 08:26:07 -0700 (MST) Message-Id: <9FAC96AF-883D-4E41-A115-7FA0FF6A18A4@lanl.gov> From: Craig Rasmussen To: WG5 In-Reply-To: <20081208024650.76234C4596C@www2.open-std.org> Content-Type: multipart/alternative; boundary=Apple-Mail-1--812819561 Mime-Version: 1.0 (Apple Message framework v929.2) Subject: Re: (j3.2006) (SC22WG5.3752) Ballot on the technical content of the TR Date: Mon, 8 Dec 2008 08:26:07 -0700 References: <20081127193527.EF00DC178D9@www2.open-std.org> <20081208024650.76234C4596C@www2.open-std.org> X-Mailer: Apple Mail (2.929.2) X-NIE-2-MailScanner-Information: Please see http://network.lanl.gov/email/virus-scan.php X-NIE-2-MailScanner: Found to be clean X-NIE-2-MailScanner-From: crasmussen@lanl.gov X-Spam-Status: No X-Proofpoint-Virus-Version: vendor=fsecure engine=4.65.7400:2.4.4,1.2.40,4.0.164 definitions=2008-12-08_10:2008-12-05,2008-12-08,2008-12-08 signatures=0 Sender: owner-sc22wg5@open-std.org Precedence: bulk --Apple-Mail-1--812819561 Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit On Dec 7, 2008, at 7:46 PM, Jim Xia wrote: > 2.) Fortran descriptors > Although Fortran descriptors are used when passing assumed-shape > arrays, pointer > arrays and allocatable arrays by many vendors, they are however not > universally used > by all vendors. Based on a survey earlier this year (by Craig > Rasmussen at J3 m183), > some vendors haven't adopted descriptors for assumed shape arrays, > pointer arrays > or allocatable arrays. The requirement in the C-interop TR may pose > implementation > difficulties for those vendors. Furthermore allowing updates on > Fortran descriptors from > C programs will likely cause safety issues and also be problematic > in consistency > check by some vendors. This becomes a sure way to introduce bugs > difficult to diagnose. I'm not sure what I said about descriptors at J3 m183 but I certainly didn't intend to give the impression that some vendors don't implement Fortran descriptors. In fact, EVERY vendor I have been able to test does. Some vendors don't include the rank as part of the descriptor, though most do. Perhaps this is what Jim is referring to. Regards, Craig --Apple-Mail-1--812819561 Content-Type: text/html; charset=US-ASCII Content-Transfer-Encoding: quoted-printable
On Dec 7, 2008, = at 7:46 PM, Jim Xia wrote:

2.) Fortran descriptors 
Although Fortran descriptors are used when passing = assumed-shape arrays, pointer 
arrays and allocatable arrays by many vendors, they = are however not universally used 
by all vendors.  Based on a survey earlier this = year (by Craig Rasmussen at J3 m183), 
some vendors haven't adopted descriptors for assumed = shape arrays, pointer arrays 
or allocatable arrays.  The requirement in the = C-interop TR may pose implementation 
difficulties for those vendors.  Furthermore = allowing updates on Fortran descriptors from 
C programs will likely cause safety issues and also = be problematic in consistency 
check by some vendors.  This becomes a sure way = to introduce bugs difficult to diagnose. 

I'm not sure what I said about descriptors at J3 m183 but I = certainly didn't intend to give the impression that some vendors don't = implement Fortran descriptors.  In fact, EVERY vendor I have been = able to test does.  Some vendors don't include the rank as part of = the descriptor, though most do.  Perhaps this is what Jim is = referring = to.

Regards,
Craig

= = --Apple-Mail-1--812819561--