Xref: news2.mv.net comp.lang.c:55135 comp.lang.c++:70786 comp.os.msdos.djgpp:794 comp.std.c:4691 gnu.g++.help:4513 gnu.gcc.help:5711
From: thads AT csn DOT net (Thad Smith)
Newsgroups: comp.lang.c,comp.lang.c++,gnu.gcc.help,gnu.g++.help,comp.os.msdos.djgpp,comp.std.c
Subject: Re: float != float and floats as return types
Date: 4 Feb 1996 22:09:25 -0700
Organization: T3 Systems
Lines: 17
Message-ID: <icPFxQ9yth8T084yn@csn.net>
References: <4ej9lb$mpc AT fu-berlin DOT de> <4elnjj$er4 AT server2 DOT rz DOT uni-leipzig DOT de>
	<4eqc7l$ugh AT godzilla DOT zeta DOT org DOT au> <4eqtu3$ddo AT der DOT twinsun DOT com>
	<TANMOY DOT 96Feb1164022 AT qcd DOT lanl DOT gov> <4ettr4$al5 AT der DOT twinsun DOT com>
	<TANMOY DOT 96Feb3105644 AT qcd DOT lanl DOT gov>
Reply-To: ThadSmith AT acm DOT org
NNTP-Posting-Host: 199.117.27.22
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp

In article <TANMOY DOT 96Feb3105644 AT qcd DOT lanl DOT gov>,
tanmoy AT qcd DOT lanl DOT gov (Tanmoy Bhattacharya) wrote:

>As you explicitly state, this is a conversion from float to float, so
>no conversion ought to take place according to the section
>quoted. This, I believe, needs to be corrected: I guess the
>possibility of extra precision was not thought of in this section. 

I recall that Plauger admitted this in one of his CUJ columns several
years ago.

As a work-around, one can cast an expression twice to trigger the
conversion:

    (float)(long double) (f1 + f2)

Thad