Re: problematic array assignment in double precision (ifort, windows)
- From: nospam@xxxxxxxxxxxxx (Richard E Maine)
- Date: Thu, 31 Aug 2006 08:40:53 -0700
Steve Lionel <steve.lionel@xxxxxxxxx> wrote:
When you call the function, do you have an explicit interface for it?
That's required since it returns an array. And is the source with that
explicit interface also compiled with /real_size:64?
I'm slightly concerned that this might be leading the OP in the wrong
direction because of a terminology confusion.
First, note that the OP said it was a module procedure, so the explicit
interface should be automatic with an appropriate USE. But then, I'd be
a lot more comfortable actually seeing everything relevant instead of
just hearing it described as being in a module.
Second. I cringe when I hear people using terminology that confuses an
interface body with an explicit interface. I know that you, Steve,
understand the distinction, but many users do not. In the early days of
f90, I recall correcting some quite august experts (including people on
what was then x3j3) who misunderstood the implications of the
distinction, so I can sure see novices having the same confusion.
When the standard says that something requires an explicit interface,
that does not mean that there has to be an interface body around
anywhere. An interface body is only one of the ways to provide an
explicit interface (and in my opinion, it is the worst way in most
cases). Talking about "compiling the source for an explicit interface"
seems likely to perpetrate the confusion.
Again, Steve, I know that you know all this. But I heard the OP say that
he had a module procedure, but then later say that he needed to go read
up on explicit interfaces. This makes me concerned that he might be
getting lead down the wrong track.
To the OP. If the procedure is indeed a module procedure, then do *NOT*
also write an interface body for it; that will cause problems rather
than solve them. I'm afraid I'm a bit short on constructive suggestions
at the moment, except that, as mentioned elsethread, a full test case
might help; there is always the possibility that the problem lies in
something that you neglected to mention because you didn't realize it
could be related.
--
Richard Maine | Good judgment comes from experience;
email: my first.last at org.domain| experience comes from bad judgment.
org: nasa, domain: gov | -- Mark Twain
.
- Follow-Ups:
- Re: problematic array assignment in double precision (ifort, windows)
- From: Steve Lionel
- Re: problematic array assignment in double precision (ifort, windows)
- References:
- problematic array assignment in double precision (ifort, windows)
- From: mvukovic
- Re: problematic array assignment in double precision (ifort, windows)
- From: Steve Lionel
- problematic array assignment in double precision (ifort, windows)
- Prev by Date: Re: Array subsetting
- Next by Date: Re: Array puzzle
- Previous by thread: Re: problematic array assignment in double precision (ifort, windows)
- Next by thread: Re: problematic array assignment in double precision (ifort, windows)
- Index(es):