This is the mail archive of the
newlib@sources.redhat.com
mailing list for the newlib project.
Re: SUSv2: Can we have MAXFLOAT in <math.h> ?
- From: "J. Johnston" <jjohnstn at redhat dot com>
- To: "H. Peter Anvin" <hpa at zytor dot com>
- Cc: Shawn Behrens <sbehrens at gmx dot li>, newlib at sources dot redhat dot com
- Date: Mon, 08 Jul 2002 11:47:49 -0400
- Subject: Re: SUSv2: Can we have MAXFLOAT in <math.h> ?
- Organization: Red Hat Inc.
- References: <5.1.0.14.2.20020707095808.015cc728@localhost> <5.1.0.14.2.20020707222321.0163eec0@localhost> <3D2912DC.9020008@zytor.com>
"H. Peter Anvin" wrote:
>
> Shawn Behrens wrote:
> > Hi,
> >
> >> Shawn Behrens wrote:
> >> > SUSv2 specifies that MAXFLOAT be part of math.h. See here:
> >> > http://www.opengroup.org/onlinepubs/007908799/xsh/math.h.html
> >> >
> >> > Could that be added, please? Like this, for example:
> >> >
> >> > #define MAXFLOAT ((float)3.40282346638528860e+38)
> >> >
> >>
> >> /* math.h */
> >>
> >> #include <limits.h>
> >> #define MAXFLOAT FLT_MAX
> >>
> >> ... presumably?
> >
> >
> > Hmm. So you'd have math.h include limits.h ? I'm not sure, is that
> > "kosher"? (When I say "I am not sure", I mean it.)
> >
> > What I am thinking about is "getting more than you bargained for" ... I
> > am entirely uncertain as to what SUSv2 or ISO-C (99?) says about pulling
> > header Y in when you include header X in your program.
> >
>
> It's apparently *not* kosher according to POSIX.1:2001/SuSv3. Perhaps
> the value can be extracted mechanically?
>
> -hpa
There should be no problem just adding the value to math.h like glibc does.
I will do so shortly.
-- Jeff J.