Re: Exceeding recl
From: Richard E Maine (nospam_at_see.signature)
Date: 03/09/05
- Next message: Herman D. Knoble: "Re: CVF to Intel USE statements"
- Previous message: Mike Zulauf: "Re: Problems installing lapack"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Wed, 09 Mar 2005 08:50:46 -0800
In article <21DEC04.14011132@skyway.usask.ca>, meek@skyway.usask.ca
wrote:
> In a previous article, Thomas Koenig <Thomas.Koenig@online.de> wrote:
> >What's people's opinion on the code below? Is it invalid? Does
> >a processor have to catch this?
> >
> > a = 3.0
> > b = -1.0
> > open (10,recl=20)
> > write (10,'(2E20.10)') a,b
> > end
> >
> Yes,but just on execution, not compile.
Do you have any support for that claim? I certainly know of none.
I certainly agree that it is desirable for a compiler to catch this (and
that it is most unlikely to be doable at compile time). I also suspect
that most compilers do catch it (during execution).
But Thomas didn't ask whether it was desirable or whether most compilers
do. He specifically asked whether a processor "has" to. I know of no
justification for claiming that it "has" to. If there is some, I'd like
to see the citation in the standard. Note that it is not enough to find
a citation saying that the code is invalid would not be enough to show
that a compiler has to catch it. (That would be enough to answer the "Is
it invalid?" question, but not the "Does a processor have to catch
this?" one.)
-- 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
- Next message: Herman D. Knoble: "Re: CVF to Intel USE statements"
- Previous message: Mike Zulauf: "Re: Problems installing lapack"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|